Discusión de los resultados

improving and deteriorating international competitiveness is actu- ally a matter of prices (and especially those of traded goods). Even so, the pace of aggregate nominal wage increases is a decisive factor for macroeconomic imbalances because nominal wage and price increases tend to move together, as the one drives and justifies the other: faster price increases feed higher nominal wage demands, while faster wage growth pushes up domestic firms’ costs, while at the same time creating scope on the demand side for higher prices. And it is not just wage developments in the tradable sector (often approximated by manufacturing) that are crucial. this is because manufacturing buys in many domestic services that therefore influ- ence its cost base (Horn et al., 2007). And even in the sheltered sector the pace of wage growth is important for import dynamics. 31

it is important to emphasise a related point in this context. the fact that nominal wage and price increases are closely correlated also means that real wage increases – which is what workers and the unions that represent them are ultimately interested in – are not closely linked to the pace of nominal wage growth; decisive for real wage increases is, rather, the rate of productivity growth.

these theoretical interlinkages (and their limitations) can be seen in the figures for the euro area countries collated in table 1. 31 It is surprising how many commentators focus solely on exports, forgetting that imports are just as important in determining the current account position.

Table 1: Wage, price and productivity variables, %-change 2000-2008 NW RW Pdty ULC P BE 25.6 5.8 6.4 18.0 18.7 DE 8.9 0.5 8.9 0.0 8.3 IE 56.4 25.7 10.4 41.6 24.4 EL 52.1 16.7 18.7 28.2 30.4 ES 35.9 0.4 3.9 30.8 35.4 FR 25.0 5.6 6.0 18.0 18.4 IT 27.2 4.4 1.3 25.7 21.9 LU 26.7 -4.9 1.8 24.5 33.2 NL 33.1 9.3 11.7 19.1 21.8 AT 21.0 5.7 11.8 8.2 14.5 PT 30.0 3.8 5.7 23.1 25.3 FI 30.7 17.7 13.7 14.9 11.1 EA-12 22.2 3.3 6.7 15.7 18.2

Notes: NW (‘nominal wage’) = nominal compensation per employee RW (‘real wage’) = real compensation per employee, GDP deflator32

Pdty (‘productivity’) = GDP per person employed, constant prices

ULC (‘nominal unit labour costs’) = Ratio of compensation per employee to real GDP per person employed

P (‘prices/inflation’) = GDP deflator Source: AMECO

32 The GDP deflator is used to measure prices and calculate real wages. Unlike the consumer price index this measure focuses on the prices and costs of domestic production. (See earlier footnote.) The period 2000-2008 was characterised by rising energy prices: consumer price inflation was rather higher.

looking first at the EA-12 aggregate we see the roughly parallel increase in prices and unit labour costs. real wage growth was much closer to productivity gains (although still lower) than to nominal wage growth. With some exceptions these patterns hold broadly across the countries. picking out some examples we see that despite the fact that nominal wage growth in spain was about four times the rate in Germany, the rate of real wage increases in the two countries was almost identical. Germany differs from the rest – and given its weight this affects the euro area average – in having a substantial difference between productivity and real wages and between nomi- nal unit labour costs and prices. We will return to this below: suffice it to say here that this reflected a major shift in national income away from labour and in favour of capital in this period.

summing up two important findings for nominal wages, these interlinkages imply that faster nominal wage growth to the extent that it is associated with faster price inflation does not raise real wages. Conversely, nominal wage moderation does not reduce the pace of real wage growth provided and to the extent that it is reflected in slower price inflation. if we now recall that wage bar- gains are always struck in nominal (cash) terms, we can draw some stylised conclusions for wage policy in a monetary union.

let us first consider a situation where national economies within a monetary union are ‘in balance’, implying that each has low unem- ployment, stable inflation, and small current account deficits and surpluses. then suppose that nominal wages grow at a rate equal to the sum of medium-run national labour productivity growth plus an allowance for the rate of inflation that the monetary authority considers compatible with price stability.33 _{in such a situation all}

countries experience the same rate of growth of unit labour costs and this rate is close to what the monetary authority considers 33 For a more detailed and technical exposition of this section see Watt 2007.

compatible with ‘price stability’. if, furthermore, there is no shift in national income between profits and wages, then domestic price inflation will increase at the same rate as ulCs.

For as long as these conditions hold the central bank can allow the economy to grow vigorously and keep unemployment low or drive it down. indeed, in the case of the ECb, it would be obliged to do so by its secondary mandate (to support the goals set out in the treaty), given that its primary mandate (price stability) would be assured. this remains true for all that the ECb has sought to down- play this obligation. meanwhile, real wages in each country grow in line with the medium term rate of productivity in that country and workers’ share of national income is stable.

taking medium term productivity growth helps to smooth out cyclical fluctuations. using the central bank inflation target as a guideline, rather than current price inflation, does the same, and, crucially in the light of the above discussion, prevents nominal wage and price developments in member countries diverging over time. taking national productivity growth promotes social and regional

cohesion within countries. yes, in a physical sense the rate of pro- ductivity growth differs between sectors. but the outcome outlined here – uniform increases across sectors – is compatible with that due to changes in the relative prices of goods produced by different sec- tors. indeed, this is what we see in practice: the price of, say, haircuts, rises relative to those of mass-produced widgets, while substitution between, say semi-skilled hairdressers and widget makers tends to balance their wages. meanwhile individual producers have an incentive to raise their productivity: if they beat (underperform) the sectoral average they earn higher (lower) profits. this is not the case in a stylised ‘superflexible’ wage-setting system idealised by some liberal economists and policymakers: if workers’ wages responded immediately and completely to the productivity of the individual plant there would be no incentive at all to raise productivity.

overall, this is a policy mix, with wage-setting at the centre that not only avoids macroeconomic imbalances but also maximises growth and employment opportunities and real incomes, while maintaining price stability. it does not ignore differentials of produc- tivity levels or growth, nor country’s different production and export specialisations. All countries should indeed strive to raise productiv- ity and adapt their specialisations. to the extent that productivity- enhancing policies are successful, the pace of both nominal and real wage growth can and should increase. productivity is the cloth from which the cloak of real living standards is cut.34

now let us consider the case where the starting point is one of substantial current account imbalances. if these are to be corrected35_,

the rate of nominal wage growth should be lower than indicated by the above formula in deficit countries and higher in surplus countries to bring countries back into equilibrium. this wage norm – nominal wage growth in each country equals medium term national pro- ductivity growth, plus the target inflation rate of the central bank, plus/minus a competitiveness correction in surplus/deficit countries – can be seen as the “Golden rule” of a monetary union (Watt 2010). it would be sensible to apply a cut-off point or floor to this rule, such that negative nominal wage growth (i.e. pay cuts) in deficit countries should be avoided in order to avoid the risk of cumulative deflation (as opposed to relative disinflation). to put it another way, 34 Clearly, it is a highly stylised model. Among other things it assumes high mobility of labour within a country (which to some extent in Europe is an arbitrary geographical area from an economic point of view), and limited labour mobility between countries.

35 The existence of small current account imbalances even in the long- term may well be considered an ‘equilibrium’ feature of economies in a monetary union, reflecting economic catch-up, demographic factors etc. I do not enter the debate here as to exactly when an imbalance should be considered excessive.

a nominal wage freeze would be the most severe adjustment path envisaged under such a rebalancing strategy. A somewhat higher overall inflation target would facilitate inter-country adjustment while avoiding costly deflation, by permit- ting faster wage and price increases in surplus countries

(Allsopp /Watt 2003, blanchard 2010).

We have thus outlined a model under which, if its conditions are met36_{, current account imbalances can be reduced while, in the}

short-term, maintaining output and employment as far as possible in both deficit and surplus countries. moreover, in the medium and longer term it would avoid the creation of imbalances between countries and keep the whole currency area on a balanced growth path with low inflation and high employment.

so much for theory.